An assistant answers well the questions that can be answered with the information the company gives it: prices, opening hours, address, scope of services, appointments and return terms. Questions that require a decision, negotiation or individual review should be passed to a person without guesswork.
A company thinking about adding an AI assistant to its website usually starts by asking "which questions can it answer?" The answer depends less on how "smart" the assistant is and more on what information it is given. A well-built assistant is the spoken form of the company's written information; it cannot know anything that is not in that information, and it should behave accordingly.
Questions the assistant answers well
A large share of customer questions are similar to one another, and their answers are already known. This is where the assistant is most reliable:
- Prices and packages. List prices, what each package covers, cases that incur extra charges.
- Opening hours and address. Which days you are open, holidays, how to get there, parking.
- Scope of services. The "Do you do this?" question; the work you do and do not do.
- Appointments and applications. How to book an appointment, which documents are needed; redirecting to the booking system if there is one.
- Return, cancellation and delivery terms. Whatever the company's written terms say.
What these questions have in common is this: the answer is written down in a document, a price list or a terms text. The assistant reads that text and gives a short, clear answer in the way the customer asked.
What does the assistant learn from?
The quality of the assistant's answers is limited by the quality of the information it is given. During setup, the following information is usually collected: services and prices, opening hours and address, frequently asked questions and their answers, return and cancellation terms, appointment or application steps. The clearer and more up to date the information, the clearer the answers.
When the information changes, the assistant must be updated too. A company whose prices have changed but whose assistant still quotes the old price is giving customers wrong information. That is why it should be decided from the start who will update the information after setup, and when. We explain how this workflow operates on the AI Customer Assistant page.
Which questions should be passed to a person?
There are also questions the assistant should not answer. These are usually questions that require a decision rather than information:
- Requests for an individual price, a discount or negotiation.
- Situations that need to be looked into, such as complaints, damage or delays.
- Assessments of the customer's own file or order, or of personal matters such as health or legal issues.
- Any question whose answer is not in the information provided by the company.
For these questions, the right behaviour is for the assistant not to guess, to state the situation clearly, and to take the customer's contact details and pass them on to the company. We cover the risk of wrong answers and how it is limited separately in the article if the website assistant gives a wrong answer.
Questions outside business hours
The assistant's most visible contribution is with questions that come in outside business hours. A customer who visits the website in the evening or at the weekend can get an answer without waiting until morning. If the answer to a question is not known, the customer's phone number or email is recorded and the company gets back to them on the next working day. This way, a customer who arrives outside business hours is captured rather than left unanswered and writing to another company.
Points to watch
The most common mistake when setting up an assistant is instructing it to "answer every question". The right setup is the opposite: the assistant answers only with the information it has been given, says so when it does not know, and passes the question to a person. In addition, receiving regular summaries of conversations shows the company which questions customers ask and where the information has gaps; these summaries are the most practical way to complete the assistant's information over time.
In short, the assistant takes on frequently asked questions with known answers and leaves questions that require a decision to a person. When this line is drawn clearly from the start, the assistant saves time for both customers and the team.
Frequently asked questions
Can the assistant answer every question?
No, and it should not. The assistant only answers questions that can be answered with the information the company has given it; questions not covered by that information or that require a decision are passed to a person without guesswork.
Can the assistant give prices?
It can give the list prices from the company's written price list. Requests that require a decision, such as individual prices, discounts or negotiation, are passed on to the company.
What happens when the information changes?
The assistant's information must be updated as well; otherwise it will keep giving the old information. That is why setup includes deciding from the start who will update the information and how often.
Is there a job that keeps repeating at your company? Tell us about it and we will build the tool that does it.
Tell us what you needOther articles
Turning hundreds of PDFs and documents into one table: by hand or with AI?
If there are few documents and they share the same layout, manual entry may be enough. If documents number in the hundreds, their layouts differ and new ones keep arriving, reading them with AI and turning them into a table makes more sense; but suspicious fields must be checked by a person.
25 September 2026Read →How many days does an automation project take?
Duration is set not by how much code is written but by how organized the data is and who needs to approve, and when. Even a simple task can stretch out as more approvers get involved.
23 September 2026Read →Does a small company need an AI budget?
The budget is not set by the size of the company but by how many hours a specific task wastes. A small company can also start with a single repeating task and a measurable budget.
23 September 2026Read →Who benefits from an appointment system?
An appointment system can help when customers need to reserve a time slot and arranging those slots by phone interrupts your work. First clarify service durations, shared resources and exceptions.
22 September 2026Read →What if AI makes things up?
A fluent answer is not verified information. A business tool needs more than well-written output: it needs a workflow that prevents actions based on incorrect or incomplete information.
22 September 2026Read →Can a spreadsheet with thousands of rows be entered into a system in bulk?
In most cases, yes. The hard part is not the entry itself but getting the data into the shape the system expects and verifying that every row was really saved. If the system offers no bulk route, the work goes through the screen, and that depends on the rules of that system.
21 September 2026Read →Why does preparing a quote take so long?
Writing the quote does not take long; the time goes into hunting for the price, copying and correcting an old quote, and rebuilding the layout every time. Once the price list sits in one place and the rules are written down, a quote comes out in minutes.
21 September 2026Read →Which jobs are suited to automation, and which are not?
The job best suited to automation is one that repeats often, whose rules can be written down and whose data is already digital. Work that needs judgement, is full of exceptions or has its data on paper needs another step first.
18 September 2026Read →What if the AI assistant on your site gives a wrong answer?
The risk is real, but manageable. An assistant's wrong answer usually comes from guessing about something it does not know. The solution is to draw a clear line from the start around what the assistant knows and does not know, and to have it hand over to a person when it does not know.
18 September 2026Read →Is ranking in Google the same as being mentioned by AI?
Partly. Both draw on the same foundation, but they do not reward the same thing: Google ranks a page, while AI reads several sources, writes an answer and decides for itself who to mention in it.
17 September 2026Read →Why doesn't ChatGPT mention my company?
The reason is rarely “the AI does not like us”. Usually it is a technical block, scattered company information, or the company not being mentioned anywhere else online. All three can be fixed.
17 September 2026Read →What should your site have so AI describes you correctly?
When AI describes a company, it uses the clearest sentences it finds on the site. The list below goes page by page through where those sentences should sit, how they should be written and the markup machines will read.
17 September 2026Read →What is GEO? Being visible in AI answers
GEO (Generative Engine Optimization) is the work of making sure that when ChatGPT, Gemini, Perplexity and Google's AI Overviews answer a question, they find your business, describe it correctly and cite it as a source. It does not replace SEO; it builds on top of it.
17 September 2026Read →Will an AI tool learn my company's data?
Short answer: not in a tool running through a commercial API. But that answer depends on how the setup was done, so it is worth knowing what goes where.
16 September 2026Read →When does a spreadsheet stop being enough?
Excel is not a bad tool; it handles most jobs and more. The problem is carrying on with it for months after the point where it stopped being the right fit.
16 September 2026Read →Off-the-shelf software or custom-built?
The right answer depends on how specific the work is to you. Work that everyone does the same way is cheaper off the shelf; work you do your own way is cheaper built for you.
16 September 2026Read →